- This application allows a client to use a predictive model to determine which user is more likely to have tweeted a given text
- Developed framework using Flask Python that queries the Twitter API for tweets from various users
- Implemented word2vect using a SpaCy NLP model to create embeddings from the tweet text
- Stored embedded tweet data in a SQLAlchemy Database
- Fit Scikit-Learn Logistic Regression model to tweet data to make predictions, serializing the results for online use
stevenhastings / tweetypy Goto Github PK
View Code? Open in Web Editor NEWFull-Stack application that allows client to use a predictive model to determine which user is more likely to have tweeted a given text. This project covers everything from API's to Predictive Modeling, SQLAlchemy database storage, Flask, along with other full-stack components. In the end it is deployed for online usage using Heroku.
License: Other